nemo_automodel.components.models.glm4_moe.state_dict_adapter

Module Contents

Classes

Name	Description
`Glm4MoeStateDictAdapter`	Converts between HF GLM4-MoE checkpoints and our grouped-experts native format.

Data

logger

API

class nemo_automodel.components.models.glm4_moe.state_dict_adapter.Glm4MoeStateDictAdapter(
    config: typing.Any,
    moe_config: nemo_automodel.components.moe.config.MoEConfig,
    backend: nemo_automodel.components.models.common.BackendConfig,
    dtype: torch.dtype = torch.float32
)

Bases: MoESplitExpertsStateDictMixin, StateDictAdapter

Converts between HF GLM4-MoE checkpoints and our grouped-experts native format.

GLM4-MoE HF experts use keys: model.layers.{L}.mlp.experts.{E}.gate_proj.weight model.layers.{L}.mlp.experts.{E}.up_proj.weight model.layers.{L}.mlp.experts.{E}.down_proj.weight model.layers.{L}.mlp.shared_experts.gate_proj.weight model.layers.{L}.mlp.shared_experts.up_proj.weight model.layers.{L}.mlp.shared_experts.down_proj.weight

Our native format groups them into: model.layers.{L}.mlp.experts.gate_and_up_projs # [n_experts, dim, 2*moe_inter_dim] model.layers.{L}.mlp.experts.down_projs # [n_experts, moe_inter_dim, dim] model.layers.{L}.mlp.shared_expert.gate_proj.weight model.layers.{L}.mlp.shared_expert.up_proj.weight model.layers.{L}.mlp.shared_expert.down_proj.weight

nemo_automodel.components.models.glm4_moe.state_dict_adapter.Glm4MoeStateDictAdapter.convert_single_tensor_to_hf(
    fqn: str,
    tensor: typing.Any,
    kwargs = {}
) -> list[tuple[str, typing.Any]]

Convert a single tensor from native format to HuggingFace format.

Parameters:

fqn

str

Fully qualified name of the tensor in native format

tensor

Any

The tensor to convert

**kwargs

Defaults to {}

Additional arguments for conversion

Returns: list[tuple[str, Any]]

List of (fqn, tensor) tuples in HuggingFace format

nemo_automodel.components.models.glm4_moe.state_dict_adapter.Glm4MoeStateDictAdapter.from_hf(
    hf_state_dict: dict[str, typing.Any],
    device_mesh: typing.Optional[torch.distributed.device_mesh.DeviceMesh] = None,
    kwargs = {}
) -> dict[str, typing.Any]

nemo_automodel.components.models.glm4_moe.state_dict_adapter.Glm4MoeStateDictAdapter.to_hf(
    state_dict: dict[str, typing.Any],
    exclude_key_regex: str | None = None,
    quantization: bool = False,
    kwargs = {}
) -> dict[str, typing.Any]

nemo_automodel.components.models.glm4_moe.state_dict_adapter.logger = logging.getLogger(__name__)